dret
commented
5 years ago review response:
hello jan.
thanks for the review!
On 2018-11-20 07:12, Jan Algermissen wrote:
> To me the draft looks fine.
that's good to hear.
> Tow remarks / questions
> - Maybe it is useful to add information about any relationshipt to Content-Location response header if present?
afaict, Content-Location in a response indicates the URI of a resource that, at the time of the request, would return the same response as the one in the HTTP exchange. it is not concerned with making any statements about future URIs or alternative ones.
for this reason, i don't think that Sunset does have any relationship with Content-Location. before the sunset happens, Content-Location can be used in the usual way at the original URI. should the resource move to a new identity, Content-Location can be used at the new URI. in both cases, Content-Location will be used as usual.
> - Why is the sunset information reference not simply an additional, optional field of the Sunset header itself?
the idea of the Sunset header is to present once an upcoming Sunset is communicated, i.e. it is about announcing the sunset itself.
the idea of the sunset link relation type is to link to information about sunset management, e.g. a resource explaining how far in advance sunsets will be announced. this resource can be available all the time, not only when the upcoming sunset is known and announced through a Sunset header in responses. the relation can be used in formats such as the JSON home format, providing information about an API as a whole.
thanks for the feedback and cheers,
dret.
--
erik wilde | mailto:erik.wilde@dret.net |
| http://dret.net/netdret |
| http://twitter.com/dret |
expert review by @algermissen (thanks!):
To me the draft looks fine.
Tow remarks / questions